A neural fuzzy controller (NFC)-based speed controller for the sensorless permanent magnet synchronous motor (PMSM) drive control system is realized in this paper. The NFC is a fuzzy logic controller (FLC), which adjusts the RBFNN-based (radial basis function neural network) parameter by adapting the dynamic system characteristics. For sensorless PMSM drive, the integration of sliding mode observer (SMO) and phase-locked loop (PLL) is executed to estimate the rotor position and speed. To eliminate the initial rotor position estimation and overcome the conventional PLL-based position estimation error in the direction reversion transition, the I-f control strategy is applied to start up the motor and change the rotational direction effectively. The system performance was verified in various experimental conditions. The simulation and experimental results indicate that the proposed control algorithm is implemented efficiently. The motor starts up with diverse external loads, operates in a wide speed range for both positive and negative directions, and reverses the rotational direction stably. Furthermore, the system presents robustness against disturbance and tracks the command speed properly.

An observer is a crucial part of the sensorless control of a permanent magnet synchronous motor (PMSM). An observer, based on mathematical equations, depends on information regarding several parameters of the controlled motor. If the motor is replaced, then we need to know the motor parameter values and reset the observer’s parameters. This article discusses an intelligent observer that can be used for several motors with different parameters. The proposed intelligent observer was developed using machine learning methods. This observer’s core algorithm is a modified Jordan neural network. It processes Iα, Iβ, vα, and vβ to produce Sin θ and Cos θ values. It is combined with a phase-locked loop function to generate position and speed feedback information. The offline learning process is carried out using data acquired from the simulations of PMSM motors. This study used five PMSMs with different parameters, three as the learning reference sources and two as testing sources. The proposed intelligent observer was successfully used to control motors with different parameters in both simulation and experimental hardware. The average error in position estimated for the simulation was 0.0078 p.u and the error was 0.0100 p.u for the experimental realization.